Stellar Evolution and Explosions: Contributions to the Abundance Evolution in Galaxies attempt of a survey with many contributions from present and past collaborators Friedrich-Karl Thielemann Department of Physics University of Basel Switzerland How do we understand: solar system abundances.. low metallicity stars ... galactic evolution? SN 1987A s process r process Supernova Classification According to Spectra from Turatto (2003) and Turatto et al. (2007) Brief Summary of Burning Stages (Major Reactions) 1. Hydrogen Burning T = (1-4)x107K pp-cycles -> 1H(p,e+)2H CNO-cycle -> slowest reaction 14N(p,)15O 2. Helium Burning T=(1-2)x108K 4He+4He ⇔ 8Be 8Be(,)12C[(,)16O] 14N(,)18F(+)18O(,)22Ne(,n)25Mg (n-source, alternatively 12C((,pn)16O) 3. Carbon Burning T=(6-8)x108K ongoing 12C(12C,)20Ne 23Na(p,)20Ne measurements of 12 12 23 23 24 C( C,p) Na Na(p, ) Mg key fusion 4. Neon Burning T=(1.2-1.4)x109K reactions at low 20Ne(,)16O energies 20Ne(,)24Mg[(,)28Si] 30kT = 4MeV 5. Oxygen Burning T=(1.5-2.2)x109K 16O(16O,)28Si 31P(p,)28Si ......,p)31P ...,n)31S(+)31P 31P(p,)23S 6. “Silicon” Burning T=(3-4)x109K (all) photodisintegrations and capture reactions possible ⇒ thermal (chemical) equilibrium Global Chemical (=Nuclear Statistical) Equilibrium (NSE) Astrophysical Sites Hertzsprung-Russell Diagram of Stellar Evolution from Iben, showing as end stages white dwarfs and core collapse (supernovae/neutron stars, black holes, hypernovae, GRBs?) main sequence influence of reaction cross sections, e-capture in late burning stages, metallicity, rotation, magnetic fields, stellar winds on final outcome Explosive Burning C O C Ne Si typical explosive burning process timescale order of seconds: fusion reactions (He, C, O) density dependent (He quadratic, C,O linear) photodisintegrations (Ne, Si) not density dependent Explosive Si-Burning 5 Explosive Burning above a critical temperature destroys (photodisintegrates) all nuclei and (re-)builds them up during the expansion. Dependent on density, the full NSE is maintained and leads to only Fe-group nuclei (normal freeze-out) or the reactions linking 4He to C and beyond freeze out earlier (alpha-rich freeze-out). Quasi-Equilibrium (QSE) full NSE is not attained, but there exist equilibrium groups around 28Si, 56Ni and n,p,4He, which are separated by slow reactions Sample Calculations from B.S. Meyer and Hix and Thielemann Hix, Parete-Koon, Freiburghaus, FKT (2007, advantages in 3D nucleosynth.) s-, r- and p-Process F. Käppeler P. Möller Wind Losses During Stellar Evolution (Effects of Rotation) Stellar yields divided by the initial mass as a function of the initial mass for non–rotating (left) and rotating (right) models at solar metallicity (Hirschi et al. 2005) Effect of fast Rotation on Stellar Evolution and Wind Losses Spite et al. 2006 Predicted evolution of 60 Msol PopIII star with 52 or 65% critical rotation (Meynet et al. 2007). Evolution of 12C/13C ratio for stellar yields without or with the inclusion of fast rotators for metallicities below Z = 10−5 solid line/dashed line (Chiappini et al. 2009), also producing primary N and increasing N/O and C/O (Hirschi et al. 2008) s-Process (neutron) Sources Core burning of massive stars (weak s-process) 1. Helium Burning T=(1-2)x108K 4He+4He ⇔ 8Be 8Be(,)12C[(,)16O] 14N(,)18F(+)18O(,)22Ne(,n)25Mg 2. Carbon Burning T=(6-8)x108K 12C(12C,)20Ne 12C(p,)13N(+)13C 12C(12C,p)23Na 13C(,n)16O protons as well as alphas are not existing intrinsically in C-burning, as destroyed in prior H-burning and He-burning. They come from the C-fusion reaction He-shell flashes in AGB stars (strong s-process) protons are mixed in from the H-shell and produce 13C (as in 2. above), but the latter can react with the full He-abundance in He-burning and produce a strong neutron source. in low and intermediate mass stars the H- and He-shells are located at small distances. They do not burn in a constant fashion. If the H-burning zone is on, it creates He fuel. After sufficient He is produced, He is ignited in an unburned He-rich zone (at sufficient densities and temperatures). The burning is not stable, the amount of energy created in a shallow zone is not sufficient to lift the overlaying H-shell which would cause expansion + cooling, i.e. steady burning. Instead He-burning, being dependent on the density squared, burnes almost explosively (flash), causing then a stronger expansion which even stops H- burning in the H-shell. This behavior repeats in recurrent flashes. H is mixed into the unburned He fuel. Schematic structure of an AGB star (not to scale) H-rich H-burning convective shell envelope He-burning He shell intershell C-O core Dredge-up Flash-driven intershell convection Observations of post-AGB stars, indicating the intrinsic pollution due to strong s-processing Gallino et al. (2008) The s-process is secondary process (capturing neutrons on pre- existing Fe-group nuclei). A similar neutron exposure on smaller amounts of Fe-seeds leads to stronger production of the heaviest s-nuclei (so-called lead stars). the full process of multi-D mixing is not fully understood yet (resolution and 3D), thus the mixing efficieny is introduced by a paramter (here ST*fac) each star shows a specific stage of s-processing, i.e. we have no overall agreement with „solar“ s-process abundances in a single star. Solar s-abundances are only obtained via integrating over an IMF and over galactic evolution with increasing metallicity Type Ia Supernovae from Accretion in Binary Stellar Systems binary systems with accretion onto one compact object can lead (depending on accretion rate) to explosive events with thermonuclear runaway (under electron-degenerate conditions) - white dwarfs (novae, type Ia supernovae) - neutron stars (type I X-ray bursts, superbursts?) Back of the Envelope SN Ia e.g. W7 (Nomoto, Thielemann, Yokoi 1984); delayed detonations (Khokhlov, Höflich, Müller; Woosley et al.) P. Höflich in spherically symmetric models description of the burning front propagation (with hydrodynamic instabilities) determines outcome! W7 (Nomoto, Thielemann, Yokoi 1984) 56Ni N S 56Fe E 54Fe a deflagration (subsonic burning front) with a propagation speed related to a mixing length in time-dependent mixing length theory of 0.7 times the pressure scale height. Neutronization via electron capture (high Fermi energies at central densities) 56Ni 54Fe, 58Ni 56Fe 50Ti,54Cr (a) Test for influence of new shell model electron capture rates (including pf- shell Langanke, Martinez-Pinedo 2003) (b) Test for burning front propagation speed (Brachwitz et al. 2001) direct influence on dominant Fe-group composition resulting from SNe Ia Ignition density determines Ye and neutron-richness of (60-70% of) Fe-group FKT et al. (2004) results of explosive C, Ne, O and Si- burning: Fe-group to alpha- elements 2/1-3/1 SNe Ia dominate Fe-group, over- abundances by more than factor 2 not permitted maximum central density 3 109 gcm-3 Future 3D Models Travaglio, Reinecke, Hillebrandt, FKT (2004, 3D nucleosynthesis with tracer particles) consistent treatment needed instead of parametrized spherical propagation, MPA Garching (Röpke et al. 2007), U. Chicago/SUNY Stony Brook (Calder et al. 2007) - distribution of ignition points uncertain - hydrodynamic instabilities determine propagation of burning - deflagration/detonation transition Multi-D Simulations Snapshots of a full 3D multi-spot-ignited simulation of a SN-Ia deflagration (first three panels). Orange tones mark the WD itself, the blue surface is the flame front. Structures from Rayleigh-Taylor instabilities (‘mushrooms’) are evident. In the bottom- right panel the density structure of the ejecta is shown (from Röpke et al. 2007). Alternatives: Gravitationally confined detonations (Plewa et al. 2004, 2007, Kasen & Plewa 2007) Zorro diagram The distribution of the main abundance groups in a sample of SNe Ia. The enclosed mass of diff erent burning products is plotted vs. ∆m15 (B). Open circles refer to stable 54Fe and 58Ni; solid circles to 56Ni, and open triangles to the sum of these. Crosses show the mass enclosed inside the layer of intermediate mass elements (IME), i.e., the total mass burned. 54Fe and 58Ni in the SN core are roughly constant over all luminosities,while 56Ni determines the luminosity and correlates with ∆m15 (B). The mass enclosed by IME's is inferred to be similar for all SNe of the sample, and the explosion energy seems constant (from Mazzali et al. 2007). Subluminous Type Ia Supernovae from Taubenberger (2008), ∆m15 (B)true of these SNe is given in parenthesis The underluminous Type Ia Supernova 2005bl (Taubenberger 2008) 56 Ni mass <0.1Msol (but large Ti and Cr abundances), large amounts of IME's, at least 50% of ejecta unburned material, metallicity in unburned matter smaller than 1/10 solar. Due to spectroscopic similarities of SN 2005bl to SNe 1991bg, 1997cn, 1998de and 1999by and that most of the latter SNe also exploded in early-type, supposedly metal-rich galaxies, this could point towards a very long-lived progenitor population for all underluminous, 91bg-like SNe Ia. Core Collapse Supernovae from Massive Stars Neutrino-driven Core Collapse Supernovae H.-T. Janka Simulations with Rotation and Magnetic Fields Liebendörfer et al (06) ,Whitehouse et al. (08) entropy and magnetic field strength 0.07s after bounce grav. wave signal should be seen with LIGO at 10kpc full solution of the core collapse SN problem probably includes: 3D, SASI, accoustic modes, MHD, rotation, collective neutrino flav. oscillations? (Duan et al. 07, Dasgupta et al. 08) “Faking” multi-D hydro with neutrinos We make use of 1D models and Liebendörfer et al. (2004) reduction factors on neutrino code AGILE/BOLTZTRAN scattering or enlargement factors in with full Boltzmann neutrino neutrino absorption in order to transport obtain typical explosion energies Black hole formation after 0.4 or 1.4s for 40Msol star?? Neutrino Emission (luminosity and mean energy) for a variety of stellar progenitors (13, 15, 20, 25, 30, 35, 40 Msun) by Liebendörfer et al.